8/18 Wisconsin tornadoes

[Shane Adams]

Shane have you seen a change in the quality of new home construction?

I've only been around newly-built homes a few years, so I can't make a comparison to older homes. However I can confirm that the builders we follow DO attach the frame to the foundation. I'm fairly certain all new homes are built that way (secured to foundation), at least in Norman.

 

[Joe Nield]

Are homes really built stronger today? I don't have any facts to back this up, but I think builders cut more corners now than ever.

I would venture to say absolutely not. I think home construction quality probably peaked sometime in the mid 20th century, although I, too, have no facts to back that up.

A subdivision has been gradually being constructed across the street from my parents' house for the last 10 or 12 years. They seriously throw these houses up in a few weeks. I call them tornado fodder.

 

[Joe Nield]

I'm fairly certain all new homes are built that way (secured to foundation), at least in Norman.

I was assisting a symposium on the Fujita scale at the AMS conference in Long Beach, CA back in 2003 when Tim Marshall gave a presentation on construction quality concerns as they relate to the F-scale. He found that, even in the hardest hit areas in Moore, the same corners were being cut at an alarming rate among new homes being constructed.

Apparently lessons learned are no match for dollars saved in the short term.

 

[Gabe Garfield]

Studies have shown that updraft intensity actually decreases around the time of tornadogenesis. This makes sense considering the mesocyclone is typically going through occlusion. The occlusion cuts off the updraft, weakening it... often (for cyclic supercells), a new UD develops ahead of the surging RFD to the southeast of the primary circulation.

I guess I should do better to define what I mean by "updraft". When I say updraft, I'm not referring to the updraft of the entire storm, but rather, the upward motion in a tornado itself.

Anecdotally speaking, I can't possibly imagine that the upward motion in the Andover, KS tornado (in the vicinity of the developing tornado) was greater before tornadogenesis than after. If you watch the video when it was hitting the Golden Spur mobile home park, cloud elements rise from near the surface and hit the ambient cloud base in just around 1 second (which indicates an extremely strong updraft).

More scientifically speaking, the presence of a downdraft does not preclude the presence of a stronger updraft. Rasmussen et al. hypothesize that a positively buoyant rear-flank downdraft will return to the center of circulation when it hits the ground. Thus, it will actually enhance the updraft in the vicinity of the developing tornado (as well as tighten the circulation which will also lead to a stronger updraft).

Gabe

 

[Bob Hartig]

Now, I'd REALLY like to know if Fujita developed the F-scale based on a derivation of probable damage from particular wind speeds, or a derivation of the probably wind speed for given damage.

That kind of expresses my own question: What exactly was Fujita after when he developed his scale? From what I've read, I doubt he was concerned primarily with rating damage for its own sake--otherwise, why even bother to assign a range of wind speeds to different points on the F scale? As somebody has pointed out, a flattened house is a flattened house. If, however, his primary concern was to extrapolate tornado intensity from the damage, then correlating damage to wind speeds makes perfect sense. And of course, that's exactly what the Fujita scale does.

Obviously a lot of complications and nuances have emerged with the F-scale, and it's interesting to see these things getting hashed out in this format and elsewhere. But in the midst of all that, I think it would pay to consider the good doctor's raison d'etre for conceiving his scale. He hoped to provide some measuring stick, however rough, for determining tornado intensity. I wonder whether Fujita himself would agree that his scale is purely a damage rating, or whether that isn't a rather dogmatic stance toward a tool he devised in order to estimate tornado wind speeds. In other word, actual intensity is the thing, not the damage it causes, which is subject to all the caprices of home construction (including lack of any construction at all in the open prairie). So maybe it's just as legitimate to refer to "F4 winds" even when F4 damage isn't present as it is to assign an F2 rating when F4 damage has occurred. If wind speed/intensity is in fact the heart of the matter, then more than just damage probably needs to be factored in when assigning an F Scale rating. After all, we do have Doppler radar.

My two cents as an interested non-expert.

 

[Glen Romine]

The obsession wit the exact wind range given in the Fujita Scale I think is what most seeking to revise it are trying to get away from. Fujita stepped in to fill a needed void - namely trying to get a feel for the frequency of violent tornadoes relative to the seemingly more common weaker tornadoes - but not having any real way to quantify that. Some of his early survey work was of atmoic bomb blasts - and trying to understand the damage patterns it created. Of course Fujita was hoping to get a feel for the likely strength of the tornadic winds that would be required to cause the amount of damage observed. After all - there was no real way to measure them in those days - so the only real information you had to work with with the width, length, and extent of damage caused by a tornado. A weakness of this method is that you can't practically place every possible structure in a wind tunnel and expose it to the same conditions in a tornado and see what it really takes to cause the structure to fail by x amount. He had no hard data to base his estimates on - they were just a ballpark estimate given the knowledge available at the time. Much has since been learned that suggests Fujita's estimates were probably too high, even in the most pristine cases (well built home exposed to tornadic winds with no debris loading).

Several questions quickly pop to mind. What about the duration of time exposed to various wind speed intensities? Will exposure to 210 mph winds for 1.5 seconds cause more damage than 180 mph winds for 60 seconds? How do you quantify the affects of debris loading on structure failure? What about home design/construction impacts the degree of home failure? What about once the first structure fails - and then the debris from that starts impacting the next structure downstream? Any doubts that a home being pelted with flying 2x4's won't fail faster and more extensively than one impacted by air alone, regardless of the wind speed?

Considering the case at hand, It is known that most modern homes first fail by the garage doors blowing inward - and I'd bet that the neighborhood in Stoughton featured 3-car garages on every home - a common requirement in affluent Wisconson neighborhoods. Most of these homes will also have basements - so the structure is only attached along the outside edge of the home. You can easily find examples of F4 tornadoes in say Kansas where a home with a basement has the subfloor blown away - exposing the entire basement. I haven't seen one example of this from the Stoughton WI tornado - so you are then reliant on the sub floor attachment being equivalent to a slab if you want to call removal of interior walls, but not the subfloor as F4.

Regardless, returning to the discussion on Fujita scale, it is a damage scale - not a wind scale, regardless of what we hope to gain from it at the end of the day in terms of understanding of the frequency of extreme winds. Even if you had an exact wind measurement from the Stoughton tornado and it gave you a wind speed at ground level of 210 mph, it would be wrong to classify the tornado as an F4 if no damage equivalent to the F4 classification was found. Why, you undoubtedly ask? Because the entire climatology we currently have is based on the damage estimates - and if you change how tornadoes are classified to a new standard, such as measured wind speed, then you just made the entire tornado database worthless because it now no longer has consistent classification. Of course the system is severely flawed - nobody is questioning that, but it is the system we currently have in place, so like it or not it is what it is. Really, it is vastly better than the system we had before - which was none at all.

Glen

 

[Scott Olson]

Regardless, returning to the discussion on Fujita scale, it is a damage scale - not a wind scale, regardless of what we hope to gain from it at the end of the day in terms of understanding of the frequency of extreme winds. Even if you had an exact wind measurement from the Stoughton tornado and it gave you a wind speed at ground level of 210 mph, it would be wrong to classify the tornado as an F4 if no damage equivalent to the F4 classification was found. Why, you undoubtedly ask? Because the entire climatology we currently have is based on the damage estimates - and if you change how tornadoes are classified to a new standard, such as measured wind speed, then you just made the entire tornado database worthless because it now no longer has consistent classification. Of course the system is severely flawed - nobody is questioning that, but it is the system we currently have in place, so like it or not it is what it is. Really, it is vastly better than the system we had before - which was none at all.

Glen

I certainly don't agree that our climo database would become useless. While the system for classification has been consistent the actual rating hasn't been. It's become more accurate but is quite a subjective system. Observed wind speeds and DOW recorded wind sppeds HAVE been consistent in all but a case or two with the cooresponding wind speed associated with the Fujita scale. It seems that some damage assessments have been attaching a wind-speed as well as a Fujita rating. Just because it's all we have doesn't mean we shouldn't discuss it or even work to improve it. I support an ehnaced Fujita scale that cooreleates damage/wind speed more accurately.

 

[Aaron Kennedy]

I think the original poster was trying to relate the storm scale updraft to tornado intensity hence my comments... we're obviously talking about seperate entities Gabe

Aaron

 

[Glen Romine]

I certainly don't agree that our climo database would become useless.

Well, I guess we are all welcome to our personal opinions, but good science doesn't allow you to change the methodology of classification midstream, even for one data point. This would be akin to throwing out the data points that don't fit with your idea, but here we are adding data points to the side we want to have them in. Sure, it would be nice to know how frequently tornadoes having winds speeds greater than x speed occur over a given interval of time (for a number of reasons) - but again we only have a handful of measurements - many of which remotely sensed and NOT at the ground. Photogrammetry methods of wind speed estimation have clearly shown the horizontal wind speeds are not uniform with height in a tornado - so at what height do you decide is corect for classifying tornado intensity? Just about any method you can come up with there is going to be problems. Subjectiveness is difficult to avoid, but that was part of the reason for the formation of the QRT to try and preserve at least the high end classifications.

Someday a better system will undoubtedly come along - but for now we have a decent estimate of how often a tornado will cause x amount of damage, and would benfit from letting go of the idea of a wind intensity climatology anytime in the near future.

 

[Scott Olson]

I certainly don't agree that our climo database would become useless.

Well, I guess we are all welcome to our personal opinions, but good science doesn't allow you to change the methodology of classification midstream, even for one data point. This would be akin to throwing out the data points that don't fit with your idea, but here we are adding data points to the side we want to have them in. Sure, it would be nice to know how frequently tornadoes having winds speeds greater than x speed occur over a given interval of time (for a number of reasons) - but again we only have a handful of measurements - many of which remotely sensed and NOT at the ground. Photogrammetry methods of wind speed estimation have clearly shown the horizontal wind speeds are not uniform with height in a tornado - so at what height do you decide is corect for classifying tornado intensity? Just about any method you can come up with there is going to be problems. Subjectiveness is difficult to avoid, but that was part of the reason for the formation of the QRT to try and preserve at least the high end classifications.

Someday a better system will undoubtedly come along - but for now we have a decent estimate of how often a tornado will cause x amount of damage, and would benfit from letting go of the idea of a wind intensity climatology anytime in the near future.

I think good science requires us to refine methodology as technology and knowledge permits. As for height, 10m seems to be a well accepted standard but I totally acknowledge that any reduction formula is problematic and similar problems arise. As for the QRT from what I understand the QRT isn't even funded anymore.

-Scott.

 

[Justin Turcotte]

Our pre-Fujita scale tornado climo was developed to satisfy the needs of the Nuclear Regulatory Commission. More or less to develop a some sort of threat assessment for nuclear power plants. Aruably the most significant flaw with the F-scale is the lack of damage markers along the full extent of the tornado path. This is particularly problematic in the Plains where "well-built" structures are scarce. In my opinion it is impossible to develop a threat assessment without quantifying this bias. I hope to have this completed as a portion of my MS thesis this winter. Here is a data sample from 1950-2001 for the Norman CWA. In some CWAs the probable number of "potential" F2+ intensity tornadoes is nearly five fold higher than current records indicate. Just beacuse some county in the TX Panhandle doesn't have a high frequency of strong tornadoes based on current climo doesn't make it a prime location for a power plant, missile silo, oil refinery, or other critical infrastructure.

Back to the Stoughton tornado... On some news video I saw some houses almost leveled. The homes still had partial walls standing indicating F3 damage (and giving some indication the home was built reasonably well). If a house is completely reduced to the foundation then it deserves an F4 rating. I havn't seen an image of a home with this type of damge from the event just yet.

 

[Andy Wehrle]

Video up:

http://sphs.angeltowns.net/movies/August_1..._18_tornado.wmv

Caution: Do not ATTEMPT to view on dial-up or otherwise slow connection.

BTW, my home barometer measured 29.28" (about 991 mb) at about 11 PM on Thursday night. That's some low pressure system considering this is August and not April or November. And who says there aren't dynamic systems in the summer?

 

[Alex Lamers]

For those of you interested in the tornadoes and tornado count for the state of Wisconsin, the MKX PNS has been updated...

NEW TORNADOES AND INDIVIDUAL TORNADO INFO...
http://kamala.cod.edu/offs/KMKX/0508242019.nous43.html

RECORDS AND OVERALL COUNT...
http://kamala.cod.edu/offs/KMKX/0508242052.nous43.html

We are now at 26 tornadoes for the event after finding two more tornadoes...one in Eastern Dane and one more in Southern Jefferson. In addition the Stoughton tornado track was extended east into Far Southwest Jefferson County, although only F1 damage was noted in Jefferson County segment of track.

...Alex Lamers...